|Publication number||US8218080 B2|
|Application number||US 11/448,961|
|Publication date||Jul 10, 2012|
|Filing date||Jun 6, 2006|
|Priority date||Dec 5, 2005|
|Also published as||US20070126884|
|Publication number||11448961, 448961, US 8218080 B2, US 8218080B2, US-B2-8218080, US8218080 B2, US8218080B2|
|Inventors||Ning Xu, Sangkeun Lee, Yeong-Taeg Kim|
|Original Assignee||Samsung Electronics Co., Ltd.|
|Export Citation||BiBTeX, EndNote, RefMan|
|Patent Citations (33), Non-Patent Citations (9), Referenced by (5), Classifications (24), Legal Events (1)|
|External Links: USPTO, USPTO Assignment, Espacenet|
This application claims priority, under 35 U.S.C. 119(e), of U.S. provisional patent application Ser. No. 60/742,704, filed on Dec. 5, 2005, incorporated herein by reference in its entirety.
The present invention relates to applications and systems for televisions that have a digital video camera attached, and in particular to personal viewing settings, parental control and energy saving controls of the television.
There have been many research achievements in vision technologies and some of them have become feasible for practical applications, such as face detection and recognition. At the same time, digital video cameras, especially the low resolution Web cameras (webcams), are made very cheap and have become largely available for daily applications in the price aspect.
Digital television industry will benefit from these two facts by attempting connecting a TV set to a video camera. The challenge is in developing systems and applications based on the vision technology achievements. There is, therefore, a need for new systems and applications that combine television together with a digital video camera.
An object of the present invention is to provide new systems and applications that combine television together with a digital video camera. In one embodiment, the present invention provides systems and related methods and applications for using a digital video camera together with a television set. The present invention addresses components of the new systems that combine television and video camera and addresses new applications and corresponding methods that improve the performance of a television with the help of live video feed from the digital video camera.
With the attached video camera, the television applies face detection and recognition techniques to find out who the viewer is and automatically changes to the viewer's favorite settings and/or apply the preset parental control for this viewer. The face detection output could also help the energy saving controls of the television.
These and other features, aspects and advantages of the present invention will become understood with reference to the following description, appended claims and accompanying figures.
Different ways of integrating the above components are contemplated by the present invention. In one example, the camera 110, the controller module 108, the parameter setting module 104 and the display 102 are integrated (embedded) in a television set. In another example, the controller module 108, the parameter setting module 104 and the television display 102 are integrated in a TV set, and the camera 110 is then connected to the controller module 108. In another example, the television display 102 is a common TV set, but connected to a set-top box into which the controller module 108, the parameter setting module 104 and the camera 110 are integrated. Yet in another example, the television display 102 connects to a set-top box embeds the parameter setting module 104, the controller module 108, to which the camera 110 is then connected. Other ways of integrating/embedding the above components are possible and contemplated by the present invention, and the example system block diagram.
The controller module 108 further generates control signals to the parameter setting module 104, to change the current settings of the television display 102. The parameter setting module 104 further receives input signals from the remote control 106 and generates setting signals to the television display 102. The setting signals include whether to show a live video stream from the camera 110, in a PIP (picture in picture) mode or full screen mode, whether to show the input TV video signals connected directly to the Television with the current settings, or limit the input channels, etc.
The camera 100 may have different resolution and frame rates, and can be e.g. an infrared camera. The video captured by the digital video camera 110 is directly sent to the controller module 108 which, based on need/command, transforms the video format to one of the formats the television display 102 can render. The controller module 108 is able to output a control signal to switch on and off the video camera 110. If the digital video camera 110 has zooming or panning functionality, the controller module 108 is also capable of output the corresponding control signals to control these functions.
The image/video processing module 112 preprocesses the video signals from the video camera 110, by for example, changing the video resolution and frame rate so that the television display 102 can display the video signal from a webcam 110. The input video signal from the camera 110 is also processed by module 112 before sending to the face detection module 114 for detecting faces in the video frames from the camera 110.
The face detection module 114 outputs the location and size of the face(s) detected. For face detection, some pre-trained data is needed, which is stored in the storage module 111 of the controller module 108. The output of face detection module 114 can be directly sent to the decision making module 120 to select energy-saving functions such as e.g. automatic power-off. The output from face detection module 114 can also be provided to the face registration and training module 116 which is activated by signal from module 104 for face registration for new users/viewers.
The new training faces are stored in the storage module 111 and all the training faces are used for a training process which outputs some parameters (i.e., the data needed for face recognition, such as like for thresholds, etc.), for the face recognition module 118. These parameters are again stored in the storage 111. The output from the face detection module 114 can also be provided to the face recognition module 118 which based on the parameters stored in the storage 111, generates a face identification (Face ID) for the decision making module 120.
The decision making module 120 controls the video camera 110 based on input from the parameter setting module 104, and outputs personal settings to the parameter setting module 114 based on the Face ID and the pre-stored settings in the storage 111. An advanced parental control function can also be turned on through the remote control so that the decision making module 120 records/logs a user's channel surfing activity into the storage 111 and outputs the surfing activity records to the television display 102 for review.
The parameter setting module 104 accepts input from the controller module 108 change the current settings of the television display 102. The parameter setting module 104 can also accept user commands from the remote control 106 for parameter settings, and also transfer some control signals from the remote control 106 to the controller module 108, for example, switching on/off the video camera 110.
The remote control 106 is used by the user to command the various modules 102, 104, 108 and 110 in
Another function of the remote control 106 is face registration mode. In this mode, the face registration and training module 116 adds a face detected to a database, wherein the new user's name can be edited via the remote control 106.
Many approaches for face detection and recognition exist, and any one of such approaches can be implemented in the controller module 108. A brief example process for the face detection module 114 is shown in
Every possible face candidate, no matter the size and location, is extracted from the frame luminance component for testing. All the candidates in a scene input frame are tested by mapping to a binary value, and detected multiple overlapped faces are merged together to obtain a single output. As such, for each input frame, every possible face candidate, no matter the size and location, is extracted from the luminance component of the input image for testing (step 150). The candidate image window is first scaled to a standard size, for example, 24×24 (step 152). Therefore, there will be a 24×24=384 different grayscale values for each candidate. The 384 different grayscale values are then passed through a function Fd that inputs these grayscales I and outputs a scale value, which is then thresholded to obtain a binary result d=Fd(I) (step 154). If the result is 1, then the candidate is detected as a face, otherwise, it is not a face. The function used to map a standard size window of grayscale values to a binary range includes a set of parameters, which can be obtained offline and then stored in the storage 111.
During offline training for the parameters of Fd, we manually label a large number of faces fi, 1≦i≦Nf, and non-faces nj, 1≦j≦Nn, where Nf is the number of face samples and the Nn, is the number of non-face samples. We find a set of optimal parameters of Fd, such that the detection error for the samples is minimized, as:
where Θ is the parameter set of the function Fd. Any of the available face detection approaches can be used to obtain a function Fd together with a set of minimizing parameters.
For a real face in a video frame, there may be many candidates around this face being detected as a face. These detections have overlaps and are then merged together (in step 156) based on the overlapping to a single detection and this single detection result is output (in step 158) to face classification.
The next step can be face registration or face recognition. The TV display 102 includes an empty user list, and all the new users need to be registered. A face registration process can be started from the remote control 106 by an administrative user of the TV display 102, who will initially have access to the face registration mode through a password. After this administrative user's face is registered, no password will be needed if his face is detected by the face recognition module 118.
To register a new user, the administrative user needs to use the remote control 106 to enter the face registration mode. In this mode, the television 102 will show images of a new user directly from the video camera 110, and the user can freeze an image once a good view of the new user is captured. The new user's face is detected and marked with a box, and is then be confirmed by the administrative user through the remote control 106. After confirmation, the detected face is scaled to a standard size and then stored in the storage 111 of the controller module 108. For each new user, a number of faces need to be stored for a better recognition performance. User name is also entered through the remote control 106.
After the registration of all users, a function Fr in module 116 is trained to map from a standard size (e.g. 24×24) to a value ranging from 0 to n, assuming there are n different registered faces. The function Fr takes the grayscales I as input and outputs a category value r=Fr(I), where r=i means the candidate face is face i, and there is no match when r=0. A simple approach for face recognition module 118 involves computing the Euclidean distance from the candidate face from the stored registered faces, wherein the output category corresponds to the smallest value, if smaller than a threshold. If all distance is larger than the threshold, the output is 0.
Other faces recognition approaches can be used to train such a function Fr and its parameters Θ such that:
where c(i) is the category number of the registered face Ii, N is the total number of the registered faces, and
All the parameters needed for calculating the function Fr are stored in the storage 111 in the controller module 108.
In the regular viewing mode, other than the face registration mode, the result from the face detection module 114 is sent to the face recognition module 118. The face recognition module 118 uses the parameters stored in the storage 111 to obtain a face category number. This number (i.e., the face ID number) is used by the decision making module 120 to make further control decisions.
Using the face detection and recognition modules 114, 118, many applications can be added to the television system 100. Three example types of applications according to the present invention include: personal TV settings, parental controls and energy saving controls.
Personal Settings Functions
The personal settings control module 132 provides personal TV setting application. Based on the video captured by the video camera 110, the face detection and recognition modules 114, 118 determine the viewer(s) and send the information to the personal setting module 132. Module 132 adaptively adjust the television settings based on the viewer(s), the current settings information from input/output control module 130 and output adjusted settings information to module 130. Such settings include e.g. video settings, audio settings, channel settings, etc. The video settings include e.g. color and tint settings, brightness settings, contrast settings, gamma settings, sharpness settings, color temperature settings, etc. The audio settings include e.g. volume settings, adjusting a sound system setting based on the location of the viewer, speaker settings, audio effects settings, etc. Channel settings include e.g. enabling or disabling particular channels, loading a favorite channel set, etc.
For each registered viewer, there is a profile stored in the storage 111 of the controller module 108. When a registered viewer is detected by the face recognition module 118 as the only viewer, all the settings that changed by this viewer are recorded in the storage 111 as the current profile of the viewer. Module 130 has output signals to both storage 111 for recording commands, and parameter setting module 104. The next time when the television 102 is turned on and this viewer is the only viewer, based on signals from the module 130, the settings in the viewer's profile are loaded from the storage 111 to the parameter settings module 114 of the television by the decision making module 120. If there are multiple users detected, the personal settings will not be loaded and the new settings during this viewing period will not be recorded.
In addition, based on the videos captured from the digital video camera 110, the image/video processing module 112 (
The module 108 can further implement receiving video signal from the camera 110, detecting and recognizing particular motions in image of a person in the video signal via modules 114-118, and performing an intelligent task based on recognized motions via module 120.
In another implementation, the module 120 selectively performs: turning on the television display, turning off the television display, changing channels, tuning to a particular channel, changing television display speaker volume, selecting a preset color/sound mode, etc.
Parental Control Functions
For each new user that is registered, the administrative user (e.g., a parent) can set the accessible channels for that new user (e.g., child). By default, all the channels are accessible. With parental control module 134, the administrative user can block particular channels or select particular accessible channels. Based on the output from the video camera 110, the face detection and recognition modules 114, 118 determine who the viewer is. Under control of parental control module 134, if there is only one viewer, that viewer's accessible channels are enabled and other channels are blocked, and when multiple viewers are detected, the union of the accessible channels from all these viewers becomes accessible.
The input from module 118 to the parental control module 134 includes viewer ids. Outputs of module 134 include determined accessible channel list, recording commands, etc. Module 130 provides accessible channel list for each viewer as stored in memory, to module 134.
In addition, under control of parental control module 134, the administrative user can set whether to record the channel surfing activities for each viewer. If this is set, when a viewer is detected, the channel surfing activities of the viewer are recorded (via a command from module 134 through the input/output control module 130) in the storage ill, and the administrative user uses the remote control 106 to review these activities. Those activities may include e.g. the start viewing time, end viewing time of each channel viewed by the viewer, etc.
The administrative user can also set a quota for each viewer (user). Once this is set for a viewer, and that viewer is detected as one of the viewers, his/her viewing time is counted by sub-module 134. The quota can be a daily quota, weekly quota, a one-time-viewing quota, etc. If all the detected viewers have reached their quota, the television 102 automatically powers off based on command from module 134 through input output control module 130. Daily quota and weekly quota will be reset automatically at the beginning of each day/week by module 134.
Using the remote control 104, the administrative user can also control the accessible input sources of the television 102. In example, if a DVD player is connected to a DVI-1 input of the television 102, and a game station output is connected to a HDMI interface of the television 102. Further, using the remote control 104, the administrative user can control the accessible input source for each viewer and set another time quota for each of the input sources. For example, the input source from DVD players might be disabled for one viewer, and the input source from Play Station may be subject to another time quota of usage for this viewer.
Improved Energy Saving Functions
As shown in
Screen saver mode can also be available for television 102, with the output from face detection module 114. Instead of turning down the brightness, the television 102 can be switched to a screen saver mode, with a command from decision making module 120. In screen saver mode, for instance, the television 102 can be showing the family albums stored in the storage module 111.
Module 118 signals modules 132 and 134 with the viewer ids, and module 114 signals whether there is any viewer. Sub-modules 132, 134 and 136 signal commands out through input output control module 130. All sub-modules 132, 134 and 136 interact with the storage module 111 and remote control 104 through the input/output control module 130.
While the present invention is susceptible of embodiments in many different forms, these are shown in the drawings and herein described in detail, preferred embodiments of the invention with the understanding that this description is to be considered as an exemplification of the principles of the invention and is not intended to limit the broad aspects of the invention to the embodiments illustrated. The aforementioned example architectures above according to the present invention can be implemented in many ways, such as program instructions for execution by a processor, as logic circuits, as ASIC, as firmware, etc., as is known to those skilled in the art. Therefore, the present invention is not limited to the example embodiments described herein.
The present invention has been described in considerable detail with reference to certain preferred versions thereof; however, other versions are possible. Therefore, the spirit and scope of the appended claims should not be limited to the description of the preferred versions contained herein.
|Cited Patent||Filing date||Publication date||Applicant||Title|
|US5164992 *||Nov 1, 1990||Nov 17, 1992||Massachusetts Institute Of Technology||Face recognition system|
|US5585841 *||Apr 28, 1995||Dec 17, 1996||Siemens Quantum, Inc.||Imaging system with automatic hardcopy compensation|
|US5691772 *||Jun 7, 1995||Nov 25, 1997||Nikon Corporation||White balance adjustment device|
|US5799111||Dec 10, 1993||Aug 25, 1998||D.V.P. Technologies, Ltd.||Apparatus and methods for smoothing images|
|US5819035||Oct 20, 1995||Oct 6, 1998||Matsushita Electric Industrial Co., Ltd.||Post-filter for removing ringing artifacts of DCT coding|
|US5920356||Jun 6, 1996||Jul 6, 1999||Compressions Labs, Inc.||Coding parameter adaptive transform artifact reduction process|
|US6389177||Jul 2, 1996||May 14, 2002||Apple Computer, Inc.||System and method using edge processing to remove blocking artifacts from decompressed images|
|US6643410||Jun 29, 2000||Nov 4, 2003||Eastman Kodak Company||Method of determining the extent of blocking artifacts in a digital image|
|US6795106 *||May 18, 1999||Sep 21, 2004||Intel Corporation||Method and apparatus for controlling a video camera in a video conferencing system|
|US6898321||Oct 11, 1999||May 24, 2005||Snell & Wilcox Limited||Method and apparatus for blocking effect reduction|
|US7097102 *||Jul 29, 2004||Aug 29, 2006||Symbol Technologies, Inc.||System and method for decoding optical codes read by an imager-based optical code reader|
|US7120278 *||Aug 23, 2002||Oct 10, 2006||Kabushiki Kaisha Toshiba||Person recognition apparatus|
|US7170933||Dec 13, 2002||Jan 30, 2007||International Business Machines Corporation||Method and system for objective quality assessment of image and video streams|
|US7260823 *||Oct 31, 2001||Aug 21, 2007||Prime Research Alliance E., Inc.||Profiling and identification of television viewers|
|US7630561 *||May 25, 2005||Dec 8, 2009||Sony United Kingdom Limited||Image processing|
|US7636456 *||Jan 21, 2005||Dec 22, 2009||Sony United Kingdom Limited||Selectively displaying information based on face detection|
|US7643658 *||Jan 21, 2005||Jan 5, 2010||Sony United Kingdom Limited||Display arrangement including face detection|
|US7734098 *||Jan 21, 2005||Jun 8, 2010||Canon Kabushiki Kaisha||Face detecting apparatus and method|
|US20030071908 *||Sep 17, 2002||Apr 17, 2003||Masato Sannoh||Image pickup device, automatic focusing method, automatic exposure method, electronic flash control method and computer program|
|US20050013494||Jul 18, 2003||Jan 20, 2005||Microsoft Corporation||In-loop deblocking filter|
|US20050254782 *||May 14, 2004||Nov 17, 2005||Shu-Fang Hsu||Method and device of editing video data|
|US20060251382 *||May 9, 2005||Nov 9, 2006||Microsoft Corporation||System and method for automatic video editing using object recognition|
|US20070058726||Sep 15, 2005||Mar 15, 2007||Samsung Electronics Co., Ltd.||Content-adaptive block artifact removal in spatial domain|
|US20070206871||Mar 1, 2006||Sep 6, 2007||Suhail Jalil||Enhanced image/video quality through artifact evaluation|
|US20070237241||Apr 6, 2006||Oct 11, 2007||Samsung Electronics Co., Ltd.||Estimation of block artifact strength based on edge statistics|
|US20070280552||Jun 6, 2006||Dec 6, 2007||Samsung Electronics Co., Ltd.||Method and device for measuring MPEG noise strength of compressed digital image|
|EP1168823A2||Jun 18, 2001||Jan 2, 2002||Eastman Kodak Company||A method of determining the extent of blocking artifacts in a digital image|
|KR20000033070A||Title not available|
|KR20020036867A *||Title not available|
|WO2000022834A2||Oct 11, 1999||Apr 20, 2000||Snell & Wilcox Ltd||Method and apparatus for blocking effect reduction|
|WO2003010716A2||Jul 23, 2002||Feb 6, 2003||Hewlett Packard Co||Image block classification based on entropy of pixel differences|
|WO2005060272A1||Dec 15, 2004||Jun 30, 2005||Agency Science Tech & Res||Image and video quality measurement|
|WO2005111938A2||Apr 13, 2005||Nov 24, 2005||Hewlett Packard Development Co||A system and method for estimating compression noise in images|
|1||European Search Report by the European Patent Office for European Application No. 07102176 dated Sep. 28, 2007, pp. 1-12, Berlin, Germany.|
|2||U.S. Final Office Action for U.S. Appl. No. 11/399,846 mailed Aug. 4, 2010.|
|3||U.S. Final Office Action for U.S. Appl. No. 11/448,373 mailed Aug. 4, 2010.|
|4||U.S. Final Office Action for U.S. Appl. No. 11/448,373 mailed Oct. 29, 2009.|
|5||U.S. Non-Final Office Action for U.S. Appl. No. 11/448,373 mailed Mar. 17, 2010.|
|6||U.S. Non-Final Office Action for U.S. Appl. No. 11/448,373 mailed May 13, 2009.|
|7||US Non-final Office Action for U.S. Appl. No. 11/399,846 mailed Feb. 16, 2010.|
|8||Wang, Z. et al., "Blind Measurement of Blocking Artifacts in Images," Proceedings of the 2000 International Conference on Image Processing (ICIP 2000), IEEE, Sep. 2000, vol. 3 pp. 981-984, United States.|
|9||Wang, Z. et al., "No-reference Perceptual Quality Assessment of JPEG Compressed Images," Proceedings of the 2000 International Conference on Image Processing (ICIP 2002), IEEE, Sep. 2002, vol. 2, pp. I-447-I-480, United States.|
|Citing Patent||Filing date||Publication date||Applicant||Title|
|US8817174 *||Apr 20, 2010||Aug 26, 2014||Hitachi Consumer Electronics Co., Ltd.||Information recording apparatus and power-saving method therefor|
|US20100295988 *||Apr 20, 2010||Nov 25, 2010||Hitachi Consumer Electronics Co., Ltd.||Information Recording Apparatus and Power-Saving Method Therefor|
|US20110074540 *||Oct 29, 2009||Mar 31, 2011||Hon Hai Precision Industry Co., Ltd.||Control system and method for interface of electronic device|
|US20130329971 *||Dec 8, 2011||Dec 12, 2013||Nagravision S.A.||Method and device to speed up face recognition|
|EP2916539A1 *||Mar 4, 2015||Sep 9, 2015||Samsung Electronics Co., Ltd||Display apparatus and controlling method thereof|
|International Classification||H04H1/00, H04N7/00, H04H60/45|
|Cooperative Classification||H04N21/433, G06K9/00221, H04N7/163, H04N21/4532, H04N5/44, H04H60/45, H04N5/63, H04N21/4223, H04N21/4415, H04N21/44008|
|European Classification||H04N21/44D, H04N5/63, H04H60/45, G06K9/00F, H04N21/4223, H04N7/16E2, H04N21/45M3, H04N21/433, H04N5/44, H04N21/4415|
|Jun 6, 2006||AS||Assignment|
Owner name: SAMSUNG ELECTRONICS CO., LTD., KOREA, REPUBLIC OF
Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:XU, NING;LEE, SANGKEUN;KIM, YEONG-TAEG;SIGNING DATES FROM 20060524 TO 20060525;REEL/FRAME:017961/0912